These projects take advantage of the laboratory's expertise in studying molecular interactions and molecular structure. For aspect (1) indicated above, we have developed a strategy for generating disulfide-trapped peptide/MHC-I molecules and have expressed, refolded, purified, and crystallized several of these. In particular, we have examined H2-Dd molecules in which a residue in the binding cleft has been mutated to cysteine and have refolded these with peptides containing cysteine in appropriate positions. X-ray crystallographic structures of these confirm that the disulfide trapped peptides are stably fixed in the peptide binding groove. Versions of these produced with truncated peptides indicate that these molecules are peptide-stabilized in part of the peptide binding groove but are relaxed in the positions of truncation. Preliminary binding studies indicate that these molecules should allow mapping of sites of interaction with various cellular and viral chaperones. In addition, we have performed proof-of-principle experiments demonstrating that MHC-I molecules, specifically HLA-B*18:01, H2-Dd and H2-Ld, can be engineered to generate thermally stabile molecules that are then amenable to binding and structural studies. These are expected to provide the basis of imaging reagents in the form of pMHC tetramers that will have longer shelf lives and greater specificity. For (2) above, collaborative with the Berzofsky laboratory, we explored unique peptides generated as a result of vaccine immunization of mice in which cathepsin S sites were manipulated to alter cross-presentation. Our laboratory has examined the X-ray structure of several unique peptide/H2-Dd complexes. These structures explain the differential binding of the peptides generated, and explain the behavior of their respective tetramers in identifying CD8 T cell populations evoked by vaccination. These studies are now being prepared for publication. For (3), carried out collaboratively with Shihoko Aizawa and Mitsuo Honda, explores the CD8 T cell response to immunization of mice with recombinant BCG based vaccines that encode the Ag85B protein of M. tuberculosis. Several H2-Kd-restricted Ag85B peptides have been identified and we have crystallized and determined the X-ray structures of these. These structures explain that variant peptides can bind to MHC molecules like H2-Kd in novel conformations, and offer structural insight into cross-reactive and non-cross-reactive T cell populations primed by the new vaccines. peptide/H2-Kd tetramers identify two distinct T cell populations elicited by the Ag85B vaccine. These studies have recently been submitted for publication.